Abstract
Determining evolutionarily significant units in endangered species is one of the most significant challenges facing conservation biology. Often genetic information has been used as the primary basis of recommendations for evolutionarily significant units, but these data should be evaluated carefully and used in conjunction with other information. The endangered Gila topminnow ( Poeciliopsis. o. occidentalis ) has been the subject of extensive conservation biology research and genetic investigation. We extended these data to highly variable genetic markers, examined variation in microsatellite loci, and compared it with previous measures of genetic diversity for the Gila topminnow from the four watersheds in Arizona in which they are still naturally extant. Fish from Monkey Spring were the most highly differentiated from the other populations. Overall, the amounts and patterns of genetic variation were consistent with known historical and physical differences among sites. The four watersheds are highly physically isolated from one another and differ in a number of important factors in their physical habitat, biota, and the life‐history of the topminnows. Based on these geographic patterns and the genetic results, we recommend that the four watersheds all be managed and conserved separately.